I do not have any notes for improvement. I don't have a lot of visibility on every aspect at the moment. 

The setup could be a bit easier. 

The initial configuration process could be simpler.

HPE SmartStart was valuable for HPE ProLiant DL Servers. However, I believe it has been discontinued, considering your technical awareness of current developments. There is a feature called SmartStart, which refers to the HPE SmartStart CD provided by HPE. This CD was available until 2008. Now, in the present scenario, people are migrating from Server 2008 to Server 2019. Logically, according to the manufacturer, HPE, it's stated that SmartStart is not compatible with newer operating systems like 2019 or 2016. It doesn't work. However, I have personally performed a workaround. You can start by using the SmartStart CD to install Windows Server 2008 R2 as the base operating system and then proceed with upgrading it to Windows Server 2016. I have successfully implemented this approach on several servers, approximately 20 to 30 servers. I haven't encountered any issues so far. Why doesn't it cause problems? I believe that HPE is aware of this situation. Not every organization has the budget or resources to purchase new servers once the lifecycle ends. Therefore, HPE should provide some form of correction, such as upgrades or updated utilities, that can be used with these servers. By offering backward compatibility, at least until the G7 series, and supporting operating systems like SmartStart for these environments, it would greatly benefit clients. This approach would receive positive feedback for the company's support services.

I don't have any information about other servers. I can't compare it to other products. 

I'd like the solution to be a bit more stable. Most of the servers we use are okay. However, sometimes, you need to change the parts. We'd like the solution to be more robust. 

In the past, we've had to buy some RAM and hard drives, and sometimes we find they are refurbished and not actually new, and this is a problem for us in our market. 

I would like to replace a hard drive that might be saved on the fly, on the server that I'm using currently, but I can't do that. I'm forced to power off my virtual machine, turn off the server, disconnect it from the power, replace the drive, rebuild the storage, and power it on again. 

That's a lot of downtime, and it affects the services. I know that HP provides this kind of solution to another series for a bigger budget of course. 

The product must provide time travel features. Oracle Flashback Database provides such features. If there’s an error, we can travel back in time and revert it. If someone imports a wrong request, we can ask the system to go back in time so that the system has no problem. Backup and recovery are the most important features.

The technical support has room for improvement. HPE's interactions with other vendors tend to not be very productive. For example, we use VMware and whenever we have an ELA with HPE for VMware, we get support from them. Whenever there is an issue with their peers, they just talk about each other. When it gets worse, support can become really complicated.

The documentation for HPE ProLiant DL Servers has room for improvement. It should be updated. For example, it says on the documentation that this specific product or part number is supported. Still, when you ask the HPE team, the reply is that it's not supported, so HPE should update the information on the documentation.

The price of the solution could be less expensive to be more competitive.

When you speak of any enhancements in the product, you need to shift to the tool's commercial mode. You get more features in the solution if you pour in more money.

If a customer needs some features in the solution, they will have to buy it since certain features are not loaded by default in the server. With entry-level licenses, servers are shipped to customers free of cost. Customers need to pay if they want more and more features in the solution. Every OEM solution in the market, including HPE ProLiant DL, operates similarly.

The overall cost of the product needs to be lower, as it is one of the areas of concern in the product.

My customers are tilted toward GPU users, and the mixed version of the enabled hard drives require — flash drives, SSD, NNL, and PLAS — all mixed and matched in a single box. These are requirements I've already shared with the HPE account manager. 

Customers want the tiering concept inside a box, which is currently part of the storage system, but customers require it in a server. They want the operating system over PLAS and the IO-related things on SSD.  The rest of their data is on NLSF. 

I do not believe that improvements need to be made on the hardware side. 

On the software side, I would like for the management of the server overall (including reviewing its health) to be a bit easier. Right now, we have something that sits on the server to make sure all hardware drivers and other thing are up to date. This is a pain to set up.

The licensing fees are expensive and have room for improvement.

I want to improve hyper-converged sites. We are mostly working with the DL series, however, now we are trying to work with more hyper-converged.

It would be great if HP could use a ProLiant Server to run HP-UX, the Unix build from HPE. Right now, HP-UX cannot run on the ProLiant architecture. So if HP made HP-UX compatible with ProLiant Servers, it would be pretty helpful. On the other hand, Linux is also replacing HP-UX, so maybe HP doesn't need to integrate this architecture into ProLiant. 

If you look at ProLiant's trajectory, I think the next generation will be focused on adding memory. Currently, there is a gap in the memory technology, so I'm predicting that the internal disk for the next generation of ProLiant will be RAM only. I believe this high-speed disk will be available in the next generation. I think the disk will be integrated into the RAM architecture. 

With the improved RAM architecture, the disk will be faster than the first-class disk because there are other protocols between memory and disk. If you have your disk, you look at the disk in terms of the motherboard. If you add to the processor, it would be faster. All disks in this architecture will be in the solution's memory.

In the next release, I would like to see more redundancy of the CPU. That would be great.

The cost of the solution has room for improvement.

ProLiant servers could be more scalable. For example, some models can only contain four hard drives, but many of our customers expect six. If they were to add another extendable hard drive slot, it would be easier to sell the servers to our customers. 

The servers should include OneView for intelligence provisioning.

We'd like the solution to be more energy efficient so that we need less cooling or less dependence on a cooling system. 

Its management console could be better. There should be some tools and tips added so that, if a beginner's level goes for management, you can easily find tools and tips, and help them manage everything in case of emergency before going to technical support from HP.

They are pretty solid servers, but, just like all servers, they become obsolete after some time.

They can maybe provide a console to monitor the health of the servers. There should be some kind of console to which you can log in to remotely check on the health of a server, even when the server is off.

The servers that we have are only scalable to some extent. They have got a limited amount of hard drive that you can insert. Their scalability should be improved.

There is need to improve the iLO support. There's a management part of the server that they've improved and that was a big deal.

There are concerns about the HPE team in the Southeast Asia region, suggesting a need for improvements in terms of responsiveness. There have been instances where the response time was slow, causing challenges during urgent situations. Its stability also needs to improve. 

We have been experiencing a problem with the power pack. It gets fried and then we have to do a replacement.

There is a challenge sometimes with having quick access to some of these components when you need them. It takes a lot of time to get spare parts from genuine sellers.

You need the memories for servers. They're delivered. You put them in, and the system doesn't come up, so you have to return them. For me, that has been the major challenge.

In the day-to-day running, there's always room for improvement. If you need assistance in real time, it would be helpful if there was such a room where someone could be there and help you solve the problem in real time.

The number of people you are serving is increasing by the day, and they need immediate solutions to their issues. Sometimes you have limitations as to getting solutions to some of these issues that come up, but you are dependent on the information available to you. Being the product manufacturers also, I'm sure they have a lot of solutions to the different issues depending on the feedback they've been able to get from the field all across the world, especially those using their products. I think that would go a long way in helping.

HPE ProLiant DL Servers lack scalability.

My personal opinion is that the rack-mount kits in generation four were the best. The current ones are kind of overcomplicated to mount, so I would really like to go back to how the rack mount worked when we had the G4. This would really be an improvement.

We have encountered problems. The most pressing issue is that downloads depend at present on the current period. Secondly, in respect of the iLO access license page, iLO features should be included with the server. 

The technical support could be a little faster.

I would like to see more of a hard drive base.

IT keeps on changing the versions of their products. There have historically been great differences in time than when the G server, the Gen8, nine, and now 10 generations of DL ProLiant server have been available, the time differences are much less. Specifically for G9, G10 even less than one year, now there is no way people like us will keep on upgrading servers without having proper ROI. HPE needs to allow us to replace the CTUs of a higher generation.

Some of the features I'm already seeing over there, i.e., really moving more towards a software layer that helps us to manage the infrastructure that we're purchasing from HPE. To use a competitor's term, as to what Nutanix is doing, HPE should be moving in a similar direction, which they are now, both in regards to a free product and a paid product; I'm actually very happy about that.

I can't think of any new features I would like to see. I'm really happy with it. I know that they're moving towards OneView.

It's a server. You plug it in and it does what it is supposed to do. It’s compliant with the HPE server line because they assemble all of the drivers that they install. The drivers have typically been extremely buggy, but we removed them from our builds. I would be nice if those drivers weren't causing critical system failures every other version, but it doesn't affect us anymore.

The technical support response time has room for improvement.

A black mark that we felt with HPE is that it failed within four to six months of use, even after it was addressed with HPE. 

The implementation should be simplified, it is not straightforward.

The product should be better tested before they send it to the end-user.

The stability needs improvement.

Support could be improved to be faster in responding.

It has something called WorldView. They should do some enhancement over WorldView to run an operating system as a code. Such a feature would be amazing for running multiple servers. They are already working on this, and they can enhance the existing versions for this.

It should also be cheaper. It is currently expensive.

I would say that for us, as we're moving to more of a cloud based solution, we're looking for ways that you can take machines that you can't put in a private cloud instance and how those can be managed from a single user interface.

I can't think of anything right now regarding improvements except perhaps the price. I think more involvement with us, the customer would be good too. They are doing better with that now though. We've had meetings with them and we really discussed the problems that we've come across. I'd say we're going in the right direction, that's for sure.

They should work with Microsoft to make licensing costs better. That's because the physical servers have processors with all these cores, and Microsoft now licenses Windows virtual server desktops based on the number of cores and the serving servers. So, if you got to build 10 or 20 servers on that desktop, you got to license them all with Microsoft. They should have a partnership with Microsoft to decrease the virtual licensing of servers in partnerships so that when you purchase the servers, you also get licensing for Microsoft on top of them. That's because it gets very expensive to license virtual operating systems to run on physical servers.

The solution is pretty complete, for what I have seen over the past three years. I can't recall if I've felt that any features were lacking within that time.

The technical support is pretty good, however, they could be a bit faster and more responsive.

They should improve the speed of the solution's processor. A core banking service like ours needs faster processing for our central server. We have to run an extensive process half yearly. The solution takes more than 24 to 30 hours to complete it. Ideally, it should take six to seven hours. Our industry is growing daily with an increase in the number of customers. Thus, we plan to move to a solution with better processing speed.

They should build risc processors with high speed. Additionally, they should reduce the prices as it is a significant concern for developing countries such as us.

The price of HPE ProLiant DL Servers could improve.

In DL 500 systems, the flow of air is not good enough, and some components get overheated. IBM systems have better airflow and scalability. 

We have a data center that has limited physical space. Therefore, we need to maximize the processing or compute systems that we use. In HP servers, such as DL 500, there are only four processors, whereas IBM servers, such as SR 916, have eight processors.

We don't get any support from HP because of our geographical location. We solve all problems ourselves by doing research on the internet. It would be good to get HP support.

We'd like to see better processors, better Intel processors. That's the most important for us at this moment.

Also, the disappearance of the ML150 can be a problem. So we probably have to switch to the ML110, we don't know yet. Wa are also using the DL160, so it can be a problem. And that's why we may have to switch some of our products to a workstation.

I've already seen quite a bit here at the conference that looks good. Well, one of the areas that I really am excited about is the iLo 5 (integrated lights out technology) on the HP blade technology where they're allowing a lot of the firmware flashing to take place through the iLo, which means I don't have to go through the OS, so that beats a lot of our servers.

Run Unix in a form that we can't reboot them more than twice or maybe less in a 30-minute maintenance window. The fact that I can flash a lot of the hardware from the iLo and not have to go through the OS means a far reduced outage time maintenance window, so that really helps take the pressure off us.

I would like to see the simplicity of having the option of loading with something like SmartStart instead of having to make the external connection. There are issues with the startup that make it a little cumbersome. It takes us more time. Even though hardware and support might be rated higher, considering how much time we spend setting things up, that impacts us a lot. That's where the points get taken off my rating.

I'm not a fan of their rails. I hate how those things rack, as they're just not nice.

This server is separate from the next one and I'd like it to be integrated so it becomes all-in-one.

The big challenge for us is our deliveries for certain countries, such as Australia and some areas of Asia.

We would like to see remote monitoring capabilities added. We are currently improving our automated monitoring for HPE servers using InfoSight.

When it comes to security there can always be room for improvement. If there was an option to build security into the system it would be an advantage.

The solution could improve by providing local support here in Sri Lanka, we do not have a local office.

We're going to the hyper converged area. I've seen OneView. I'm very pleased with how that interface looks. We saw SimplyView in a demonstration and Synergy. So until I get my hands on it, I don't know that I can say I'm looking for more improvements.

I looked at the technologies offered at a recent HPE conference. There are a lot of technologies that are only offered on the high-end servers. As an example, the SSD module that's embedded on the memory; that's only available on the HP DL380. We've invested quite heavily in the HP DL360 G9s, which means that I won't have access to that technology. I would like it if there's a possibility, even if you have to pay an additional license, to enable that technology on the DL360s.

I think that the main area of improvement would be in the price. The price only came down for us when we went to a dual vendor solution. We resorted to negotiating prices between HP and its competitors, and would have been nice to not have gone through that trouble to get a more competitive price.

Java version dependence for its out-of-band management. This is also an issue for other vendors, as well.

Boot-up time should be reduced. Restarting a server at any time is a stress-filled occasion, and waiting around for the server to come back on is agonizing. 10GB and 40GB ethernet connections are essential for virtualization workloads and consolidation.

I'd like to see more features in the iLO as it currently only provides a basic overview. Sometimes the part or serial numbers are missing, for example. We've had issues where power supplies or hard drives have become defective and the information gets pulled back into the error logs. We then have to physically look into the server. This is highly inconvenient and an annoyance which could be solved with more iLO features.

The ProLiant does what it needs to do. It's not entirely perfect. It could run at lower power, for instance, and it could run cooler. I guess all of that comes at a price.

Serial number tracking and trying to get that tied to my account. I have a terrible time with that with HP. I've got hundreds of servers around the country, and when I try to call in and keep the serial numbers straight and things like that. When I try to download drivers for them, the serial number doesn't come up, even though it's under extended coverage. I've not been happy with that part.

The biggest thing I don't like about ProLiant, and this will be very trivial, is the rails. Every generation, even within the same family generation, the rails are different. It's a pet peeve. It’s not a technical thing. We do rack-mounted servers. Having to deal with different rack - or rail kits - is a real pain. It's not a huge technical thing, but that's the biggest thing for me.

We're looking forward to OneView management, stuff like that to keep it all in check. But there's not really anything specific. It will provide a single pane of glass for management instead of going out to 50 different servers and configuring things so you get, set up a template, or do the different things through OneView. Just kick it off and replicate and automate what you're doing.

We honestly don’t have any problems. It’s a well-built machine, and there are no problems with the servers.

We have had issues with ProLiant's intelligent provisioning. Sometimes, the drivers do not come up, so we have to install them manually. 

I think the speed could be improved with OneView. At the moment, OpenManage is faster and has a better interface than HPE. 

The support lifecycle for the product for the Telco application should be better.

In terms of what could be improved, if I could have faster processors with less cores, that would be good. I know that they have a new series with AMD processors, but I have currently no access to them. This would be nice, because we need faster processors for rendering. It is more interesting than having a lot of cores. Overall, I do not see any particular improvement needed, because I think that they are already at the edge and ahead of the competition.

In the next release, I would like to see to more power, because the next generation will always require more computing power, but I'm sure that they are already doing their best for this.

They have a management solution, HPE OneView, but I didn't implement it because at the beginning it can be a bit difficult and we only have a small set of servers. So it was too much work for not enough return. So it could be simplified with better results to have a better return with this.

There is need to always keep trying for improving this product.

• Remote management – the iLO 3 (and 4) are still lagging behind DELL in terms of features and value for money and they still do not have a HTML5 remote console.
• Firmware reporting is inconsistent – iLO provides no way of loading new firmware for devices other than the iLO itself. iLO has a very basic disk and RAID reporting with no configuration options at all.
• Out of our many HPE Proliant servers we have had very few failures outside of the common disk failures. We’ve had a few motherboards die but only a very low percentage.

We're looking at things like UCS just now and we are looking for what would blend with our UCS deployment. I know that HPE are heavily involved in that, so it's been a lot of consultation and a lot of fun to get all of that to work together.

I’d like to see a lower price. That's not the whole 20% difference of why I gave it an 8. One of the negative things that we find is that the configuration tools are sometimes not easy to use. That contributes to the detriment of the score.

The latest generation has made good improvements in power consumption. G6, and G7, which is approximately 10-year-old technology, have some issues with power consumption.

The installation could be simplified.

I would also like to pay less for the out-of-the-box product, but everything else is okay.

The solution should work to be more scalable.

The pricing of the product needs to be better. Right now, it's quite costly.

We use, the service comes with three year warranty with next business day support. We would like to enhance it with basic service and a better level of support. It is not required in the 380 series, but it is still widely used. So, a better option would be same day support. 

I would like to see the booting process be a little bit faster. Instead of going through the circle loading image, it would be more interesting if it would show what it is doing during each stage of the loading process.

I can't see how you can fit much more into a small server, to be quite honest. If I was to be really picky, it needs to not drop the memory speed when you fill all the memory slots, which is a real niggle, but other than that, they've worked so well. They've done so much for us in cutting our hosting costs over the original four way boxes that these replaced. Their performance is absolutely phenomenal.

As these servers are constantly evolving with better technology and performance, I would have nothing to offer in the way of a wish list feature additions.

I would like the solution to improve its storage and performance. Also, I would like to replicate data between on-premises and cloud in future versions.

They should improve the solution's built-in quality.

The size can be improved by making it a little slick. Additionally, we are at a disadvantage in support for people in African countries using the product because we don't have access to the support team 24/7. The solution has several support levels, such as essential and proactive support. They can extend support to African countries.

HPE ProLiant DL Servers are customizable, but when you first deploy them, you basically have to rebuild configurations on them. This is true for large orders. But for small orders, you will be tied up with previous configurations of some other brand like Dell.

I would like to see an on-demand model. I think they are already going there. I would like to see something in which you can move your workloads to a cloud kind of offering. That would be something that would be ideal in the future.

From a pricing perspective, a consumption based model would be optimal. There are times when a complete purchase isn’t required and a short term burst of compute is needed.

I would like more advice about drivers and servers, depending on the application we are using.

I would like to see better license management with the iLO. It is very difficult to manage the licenses for the hardware, as they can easily be lost. Then it is very hard to get it back. That must be better.

We'd like to have a management platform that's not limited and that would give us a single view of all our systems. This full range of options is limited only to certain types of hardware.

Also, if you look at compatible infrastructure that was introduced recently, it will be like an image streamer in a compatible system. An image streamer won't be able to provision older types of servers.

Obviously, you will still have hardware failures, but it is not something you can easily fix. This is more about the regular hardware failures that you may get on daily basis such as a bad HDD, NIC failure or regular hardware failures that may be present due to the normal workload of the server, meaning that they have or exceeded the lifetime or they are presenting a huge workload.

Regarding the normal functionality of the server as is, the only improvement I would say is to continue with the research and the development of higher workload capacities as the customer needs are constantly growing, therefore the demand for what the product requires to do will also grow, to avoid the demand to elapse behind the improving of the technology needs to be continued. It's tied to the needs of the end users and any new features that they would require in order to better manage their organizations.

Providing areas with room for improvement is a tough one. We're running AMD right now, so we're not running the latest gen. We're running a gen 8, something DL series. I honestly can't provide areas with room for improvement until I see the next version of what that's going to be. I mean, it’d be nice if AMD got off their butts and provided a new chip for us to use, but right now it's all Intel.

I wish there was a clearer-cut way to back up the BIOS and restore other machines. That'd be really the biggest thing I could see an improvement on, which I think they did fix in gen 9 and 10.

I would like to see more standardized management. In the past, there was a lot of mixture between the software and the hardware features.

By default on the Gen9 server, they remove the screenports and USB ports in front of the servers, and you have to purchase them. I didn’t know that beforehand. Now, when I troubleshoot things I have to connect things in the back. The new design should fix that.

In the HPE ProLiant DL Servers generation nine, there was a problem when the expansion cards were installed on the server. There was only a small amount of room for the top expansion cards that hand could insert or remove the cables inside.

The solution could improve by providing more best practices from an architectural point of view. What are some recommended configurations for use, such as in a VM environment? HP provides some best practices but they are based on VMware, not on Red Hat. It would benefit for others to see the different use cases. How the HPE ProLiant DL Servers were used in the context of Red hat virtualization, and some configurations as an example to allow us to identify some points that we can improve in our platform. We could make our system more reliable and scalable.

The scalability of Gen8 models should be improved. They can also improve its price.

Being in Canada, and with exchange-rate fluctuation, it would be good if I could get a better price point.

It's always the next generation of hardware, of course: Who does the better job? You also can look at things and say, "Hey, we were going all blades. We were going with virtual connect.”, and do specific things in that way. We learned certain lessons doing that, of course.

For the next generation, we probably won't have that many blades. We will probably revert to rack-mounted servers, but have bigger servers instead of the smaller servers. That also evolves with the workloads you have. Over the period of time we typically run these systems, such as five years, there's a lot of change in what the users request from us. Of course, there are new developments. For example, before we started VDI, we said, “OK, if we want to do VDI going forward, we probably want to incorporate some GPU into that.” That would probably lead to new architecture and then we want to do other stuff like high-performance computing, as well, on that. The next generation probably would look completely different from what we have now.

As server technology moves forward, I expect HPE server solutions to be compatible with the advancements such as supporting 40+ GbE network adapter and further reducing the complexities in a network topology such as fewer cables, edge switches and simple administration.

This seems to be important to me because in today's competitive environment, the end users expect to have a high-speed SAN network, which majorly depends on the network adapter's speed and performance. In my opinion, the product must support high-speed network adapters in order to meet the customer's expectations.

I would say that it would be nice if there was a more streamlined approach to getting drivers or looking up system information from HPE's website. That's the one thing that I did appreciate on the Dell side, you put in a serial number, you see the exact build, you see all the drivers associated with that machine. That is probably the only thing that I see missing from the HPE side is a very easy to use method to get this. It does exist, there is a method to get the information, but it's not as streamlined.

I think the big thing for us, which is unique, is a dedicated HPE Cloud offering in New Zealand. It would be quite cool. I don't know if that's actually feasible or viable for HPE. Cloud is something we're interested in, but the solutions in New Zealand aren't really up to it. So you have to go to Australia. You get latency issues.

The memory has room for improvement. I want the ability to upgrade the memory to avoid crashes.

The cost of the solution is high and has room for improvement.

I would like faster internet ports included.

There's always room for improvement somewhere.

Since the management went from HPE SIM into OneView, the problem is that the older generations are not supported by OneView. The new ones are supported by OneView. We have two management platforms to manage these DL servers and it's kind of a pain in the neck. We like integration.

There's nothing that I would add to it, except maybe on the dashboard. There could be a little more analytics. I would like to see some proactive warnings about maintenance issues. If we were to have an issue, I'd like some more advanced warning. I’d like to see some alerting features. We’ve only a few issues, but I would like to see some alerting improvements.

What would make it better from my point of view is if HPE spent more time on testing with the actual built-in Red Hat Linux drivers, as opposed to always trying to say, "Use our driver."

I would like to see better support of monitoring protocols.

I think maybe just improving the manageability since we would like a single pane of glass. I think there may be a few features that my team might want to see added. We use it a lot for VMware so anymore things that they can tightly integrate with VMware would be great.

The number of CPUs and maybe administrative interfaces. Nothing that I would site as a concern now. It's the continued maturing of the platform. There's always room for growth but I can't point to more specifics.

I'm worried that with the new model, HP will get rid of the processor ILO port that enables us to have a separated management network. I don't think it'll be feasible to have some models with the ILO port and some without.

I'd also like to see faster memory chips.

I also think that there's an issue with the Smart Array firmware, which is the RAID controller. It corrupts RAID 6 so that if you have an entire setup that's based on RAID 6, the whole RAID will also be corrupted.

I find the software for the servers can sometimes be a bit flaky. Upgrades can be flaky, too. For example, you get the HP easy set-up when you want to set up one from scratch. It's not uncommon that when you have to phone HP, they say, "Well you can't use the latest version, go back two versions," or that kind of thing. I would say where they control the servers, that's where they let themselves down. They let themselves down on their software.

The only improvement would be cost. If they keep driving the costs down -- and I don't think they're expensive for what you're getting -- they'd keep everybody happy.

Sometimes, there are connectivity issues and the server is not able to connect to our SAN. One example is when we reboot from SAN storage, we face issues such as not being able to detect the SAN port. After we installed two new physical hard disks, that issue went away. It should be more stable.

A lot of where they can improve the product would be related to hot flexibility. If you want to change a DIMM, you have downtime to upgrade. It would be good if you could have hot DIMMs and hot CPUs, also for hard disk power supplies and bits. Hopefully, then there shall be no downtime. The biggest problem with our user base is downtime.

If they could put more disk array in one node or rebuild it in the same server, it would be very helpful and more powerful.

The new function, which is coming at the moment, with the online firmware updates - that's one of the problems we're struggling with, with our customers, to have downtime - these live online updates are really, really a benefit we're waiting for.

Lower price. You have to pay for some of the features after the fact.

I guess more regular automatic updates would be nice. Sometimes when you call support they'll say, "Well, what version of frameware are you on?" And you'll say, "Version X, Y, Z," and they'll say, "You need to be on A, B, C."

Boot times are relatively slow. Legacy BIOS options are tricky, but discoverable and manageable with experience.

The new rail kits are terrible. The older rail kits were great and simple. It took one person to do it. Now the new rail kits for the DL line, since the Gen-8s, are terrible. It takes more than one person to be there present in order to rack one of those systems without endangering yourself or the server by letting it fall to the ground. The older rail kits were great, the new ones are horrible.

The solution is a little bit expensive. It would be ideal if HPE could adjust its pricing to make the product more affordable.

I think that for us, we have is what we need. I'm not sure that we would like to see any new advanced features because we're not using such advanced features. I'm sure there's always places to improve but we are not using all of the advanced features that are in HPE.

Perhaps the one thing is support globally, or distribution globally. Today we distribute in the United States and, over there, we have a center that can distribute for us, but I don't think they have such in the Eastern part of the world. So that could be an area for improvement.

• More adoption of Intel's new processors
• More SSD technology
• More NVMe technology
• Keeping the price down. (It's always the price, right?)

They're very slow to boot up. And we can have issues with the driver sets for all operating systems as well. The blade enclosures themselves, though, are leading-edge technology.

I think the problem is that we're not geared up for automating the build process. We're still manually deploying servers.

HPE ProLiant DL Servers could improve its portfolio, it could be more diverse because there is a limited number of them available. Dell, IBM, or Lenovo, have a better product portfolio that is diverse.

I would like to see additional work on making the intelligent provisioning work with VMware.

I honestly can't think of any areas for improvement.

It is too expensive. Its price should be improved.

The system admin interface should have better visibility while troubleshooting a problem, even when the problem is coming from the application.

At the moment, I feel that there is more there that we could be using rather than things that are missing.

It's perfect. It could go faster all the time, but the next generation is always faster.

What is missing for me is more of the administration site. If you have a cluster of four servers, there's no way to identify if they are all a member of the same cluster.

If there were a way to tag them, then you could specify where they are located within the data center. It would make it easier, because then we could spread them out. We would not have to keep them in the same cabinet in case of a power failure.

I would like to see more ease-of-use, be able to identify everything, and group them together so that we know where they are located.

It works well for us, but they should bring the price down.

It could use less power. We want to go green and so want to use as little energy as possible.

The most improvements that could be made would not be in the hardware commodity itself, but in the firmware. There's synergy between hardware and software, so you can't view hardware in an isolated way. So the improvements I'd suggest would be in the software that runs the servers.

The biggest area of improvement I can think of is with more fiber connections. Adding more fiber connections would make them even better.

More options beyond the ROM configurations for High Power computing and ULV (Ultra Low Voltage) deployments.

They can increase the controller cache. Currently, it has a controller cache of up to 4 GB. The RAID controller card comes with a maximum of 4 GB cache in HPE servers, whereas it comes with a maximum of 8 GB cache in Dell servers. 

They can also improve the port size. HP provides a 25 GB port on the server, whereas both Dell and Cisco provide a 100 GB port. 

I would like some of the software products that they offer to be integrated. In other words, you buy the server and you get the software, instead of having to pay extra for the licenses.

Although I compare them often to the Dell servers we used to have, these HP servers could be improved with faster memory. They perform well enough, but I might pay more for faster performance.

The software doesn't allow you to view isolated hardware issues. This needs to be improved.

Better power usage monitoring would be nice.

We found that the extensions could be improved and the price could be a little lower.

I am impressed with the SimpliVity solution. I need to look a bit about this.

I would like to look at Synergy for the new servers. It will be out in June or July. The Synergy Streamer Composer is one of the powerful features that Synergy has.

The Generation 9 servers have the new UEFI firmware version instead of the traditional firmware, but we don't have that on our servers. It would be great if they could run UEFI because they have longer boot times, so there's room for improvement there.

I would like to see a single pane of glass, relayed management, and fewer tours.

They should continue to maintain the stability that is there.

Configurations very simplified. The technical support and training process Very good.

The price of this solution should be lowered to make it more competitive.

More cloud and mobile features for monitoring.

We are not looking for new features in the DL260. Instead, we are looking at Synergy.

Compared to blade servers, it could use more flexibility to expand with new features, and to connect new storage systems and so on. It’s not so easy and not always possible.